Reliability Improvement of Power Devices for Electric Vehicle Traction

Power electronics converters in the power drive train of the electrical vehicles (EVs) are more prone to failure under different operating and loading conditions. Hence, it is necessary to improve the reliability of the power semiconductor device which in-turn reduce the thermal swing and improve the performance of EV. This paper proposes an inverter reliability improvement approach in which junction temperature profile of the power modules are taken into consideration. The thermal and power loss modeling are done at different ambient temperatures ranging from 25°C to 100°C and then rain-flow algorithm is used to estimate the stress during multiple cycles at the junction of power module. Further, a closed loop controller with adaptive neuro fuzzy inference system (ANFIS) is developed to reduce the fatigue cycles at the junction which will decrease the stress or degradation on power modules and improves the lifetime of the converter. The result shows that with closed loop ANFIS control, variation in temperature stress cycles is minimized up-to a large extent.